© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 18-TRANSCODER-HP
    7. 88650
    Prev
    Next
    INDEX
    Explanations

    over

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    çĶ±ä¸ŃåĽ½
    -0.32
     Away
    -0.27
    .INSTANCE
    -0.25
    linger
    -0.25
    icles
    -0.25
    ãĥķãĥ©
    -0.25
    ovel
    -0.24
     Boots
    -0.24
    rippling
    -0.24
    uet
    -0.24
    POSITIVE LOGITS
    æŁĦ
    0.27
    å´Ķ
    0.26
    åĪ©çĶ¨
    0.26
    Platforms
    0.26
    è°
    0.25
    åħ«åŃĹ
    0.25
    å¨ĺ
    0.25
     Mrs
    0.24
    åĴ¨
    0.24
    end
    0.24
    Activations Density 11.861%

    No Known Activations